2019-05-18
It's uncanny how closely this talk follows a line of reasoning I've been working on lately. If you're copied here, we may have discussed one or more of its aspects
https://www.youtube.com/watch?v=pW-SOdj4Kkk
(Jonathan Blow is an independent video game developer, speaking here at a conference in Moscow.)
Neither he nor Musk seem to understand that the solution is not to somehow work harder at space, or software, but rather at energy.
Also, Blow doesn't seem to realize how many of his complaints are due to the design of unix, which he still seems to lionize (like every other technologist I know of, save Jaron Lanier).
I watched the Cline talk he recommends over a year ago and highly recommend it.
Has anyone read the Foundation series? It's referenced in early Yes lyrics. I haven't read it but I suppose I have to now!
I think we recently discussed the idea that people living through a collapse may not be able to form a consensus as to whether it's happening. He mentions that.
We may have also discussed the idea that increasing rates of little annoyances, rather than dramatic events, could be indicative of collapse... things like software not working as well, bank websites down more often, wrong items shipped to your house or refunds given on different items than you returned despite online shopping being almost completely automated... He covers the software angle.
Past few years I've also become aware of and studied the Late Bronze Age Collapse. He mentions it, and recommends a video I also recommend on the subject. It's especially interesting because many stories in the Old Testament refer to it. Studying it has given me a much better understanding of the bible.
He says society is always experiencing collapse (or decay), even when it's doing well. In particular, knowledge transfer between generations is lossy. This is being made salient for me now that Adric is almost an adult – it's astounding how many things he and his friends aren't aware of. Some of it isn't worth knowing. Some can be learned abstractly, but at tremendous cost in time and effort. And a large part can only be learned by direct experience, and can't be transferred at all.
He thinks systems of increasing complexity are bad because they're harder to maintain. Software is a special case because it often becomes unmaintainable on time scales as short as a few years.
He laments programmers who learned to program with a high-level language like JavaScript and therefore don't know how computers work inside. Without that knowledge, they're much more likely to write inefficient code and compound the problems of software complexity.
So those are some of the points. I think it's interesting that it was given by an indie game developer at a gaming conference. Also, it's just a well-presented talk, which are few and far between.
Should I watch the Cline talk?
Watch Blow first. He summarizes Cline's talk, which goes into a lot of detail and may only be worthwhile if you're especially interested.
Bottom line, they had rapid mail with the clay tablets, like not that far off from e-mail in terms of how people used it, and we have enough of them to know what it was like to be a merchant in the Mediterranean area in the late Bronze age. Then it all disappeared and nobody could even read them until the 1800s.
Aside: Victorian London had mail deliveries 12 times a day. So in terms of how often people can exchange correspondence, it was as good as e-mail.
We watched Miracle on 34th St this past Christmas. It was the first I'd seen it since middle school and I was amazed how well it's held up in all respects. The office technology was particularly impressive: intercoms, pneumatic tubes, etc.
I don't think I mentioned it in my Apollo article, but in the documentary, pneumatic tubes can be seen in many of the mission control workstations. End Aside
I was made to study the Old Testament a lot as a kid, in both church and school. But I never learned about the late Bronze Age collapse. I only found out about it in the last few years by studying economics. The biblical tie-in makes it doubly interesting for me.
In particular I've enjoyed looking up some biblical topic on Wikipedia and just reading that. The pages on ancient history are some of the best I've seen on Wikipedia... niche topic where experts have no better venue and nobody else really bothers them, I guess.
I learned that Jews were never enslaved in Egypt, and did not help build the pyramids. Yahweh was originally the sky god in a Levantine pantheon, and experienced powers inflation (like a comic book hero) until he became the only god.
The Old Testament stories are mostly post-collapse people telling tales of times before the collapse. It was written in the middle of the first millennium BC, which was a 'dark age' between the Bronze and Iron ages.
Though decay may be constant, it appears that growth has outpaced decay lo these many centuries. Is that changing?
GDP should just measure this. As long as real GDP change is positive, growth is outpacing decay.
Well, this is at least valid for world GDP, which no one measures carefully! I've probably done the most careful measurement of it ever, as strange as that sounds. I drew on available data like everyone else, and it's all pretty sketch. I contacted the people who compile it for the World Bank and it's a bunch of interns with no funding. The Penn World Table people aren't much better.
Economists are obsessed with national accounts and don't much care about world economic activity even in the present, much less the past. Only one guy has studied it – Angus Maddison – and he died in 2010. There's a group at the university where he worked who are trying to carry on his research, but there clearly isn't much interest and they've botched what little they've done so far. Maddison himself made lots of mistakes, as any one guy working in isolation will do. Wherever you see pre-1960 data on GDP, it's based on his work.
But yes, world real GDP growth has been positive since the industrial revolution, except for recessions of a few years each.
There have been two mega-recessions in history: the late Bronze Age collapse and the fall of Rome.
How does one measure a collapse? Is GDP a good indicator? Population? Life expectancy?
Interestingly, world population isn't known to any particular accuracy (there are estimates but we don't know how accurate they are) prior to 1950! Here's a graph showing an amalgam of available estimates 1000 BC to 1800 AD
This is population, not GDP, but per-capita GDP was trendless prior to 1800 so it's a proxy. You can see the rise and fall of Rome (and the "classical" world more broadly... big economic cycles have been synchronized across the old world since long before 1000 BC). Anyway, the dents ca. 1400 and 1650 surely left a mark. They have names (different names, depending on whether you're talking about China or Europe) and I found them on Wikipedia at one point, but it seems I didn't note them down.
You can see people were feeling mighty fine when they produced Jesus.
Do you think a collapse is coming? Is this a bad thing? Do you have idea of a Utopian society, and if so, how do you measure it?
I don't know if one's coming. Growth has definitely slowed in the West since 2008, and there are plenty of ways to observe that. But that's been offset somewhat by progress in Asia. Beijing certainly seemed upbeat when we visited.
Long-term, continuing world growth will depend on our ability to transition from fossil fuel to a higher-density energy source, in my opinion. I don't know how long we have to avoid a 'dark age' in between, but it could be less than a century.
Thing is, the transition itself requires a lot of free energy, so it gets harder as the fossil resource decays. There's plenty of methane in the ocean but getting it is hard and I don't foresee that tapping those deposits will lead to boom times. I could be wrong.
When oil's first shooting out of the ground, it's too soon, because you don't know about fission yet. When you discover fission, oil's still good so you don't feel a great need to transition. By the time oil's in decline, the transition is already hard. Like we can't build a train, or pretty much anything that requires large-scale cooperation, in our country presently. So we're unlikely to replace our entire energy infrastructure.
Actually, the transition was well underway when the 1973 oil crisis caused a recession that had the federal government cut the budget for it. Reactors were shut off and the nuclear rocket program, which had functioning rocket engines ready to go to Mars, was mothballed. So that was a direct effect, though the relationship could also apply indirectly.
Here's the robotic remote maintenance setup for the molten salt reactor, mothballed in 1973. Note the use of 3-D video.
Here's bits of a nuclear rocket engine, in a box held by the grandson of its creator
Here's the grandfather, pointing to a photo of the engine on its test stand
The grandson helped make this documentary, which I helped bankroll.
The same oil crisis killed Jodorowsky's Dune. There's a documentary about his attempt to make Dune into a movie. Ironically, Dune is about oil. And the bible. And long-term civilization stuff. The movie was finally made by David Lynch when the economy recovered in 1984.
The Xerox Alto was stillborn in 1973, and resurrected by Apple in 1984.
Many examples exist of things going dormant across this particular economic gap, which is clearly visible in the historical price of oil. In the history of automobiles, it's known as the Malaise era.
Incidentally, the Soviet Union was doing pretty well until 1973. It never got the 1984 recovery, and this recession finally killed it.
Is it possible that a collapse of the type envisioned here, with historical precedent, is superseded by other factors? Like, could extremely powerful AI or bioterrorism cause a far bigger impact on collapse before general societal decay takes effect?
I've thought a lot about what drives economic cycles and whether human decisions have any power to cause or stop them. I don't really believe they do. The cycles could be driven by changes in the quality of the energy supply, but it's hard to imagine conclusive evidence for that one way or the other (beyond the sort of thing I just sketched above). It's hard to prove any causative influence on history because you can't repeat history to test it.
If I'm right, the first thing AI will do when it takes over is get fission going. Can't happen soon enough for me. People will complain all day long about the total incompetence or even criminal nature of every human government in recorded history, but when it comes to the prospect of an entity with hypothetically superior intelligence taking over, they want to prevent it!